> ## Documentation Index
> Fetch the complete documentation index at: https://api.finevoice.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# MCP Server

> Connect AI clients to FineVoice's voice, music, and audio processing capabilities via the Model Context Protocol (MCP)

## Overview

FineVoice provides a [Model Context Protocol (MCP)](https://modelcontextprotocol.io) Server that lets MCP-compatible AI clients — such as Claude Desktop, Cursor, and VS Code — call the FineVoice API directly, without writing any custom integration code.

The MCP Server exposes largely the same capabilities as the [REST API](/api-reference/introduction), including text-to-speech, voice conversion, speech-to-text, podcast generation, sound effect generation, audio separation, voice management, voice cloning, music generation, and audio enhancement.

**Endpoint**

```
https://mcp.finevoice.ai/v1/mcp
```

**Transport**: Streamable HTTP

***

## Authentication

The MCP Server uses the same authentication method as the REST API — a Bearer token passed in the `Authorization` header:

```http theme={null}
Authorization: Bearer YOUR_API_KEY
```

When configuring your MCP client, add this header to the corresponding `headers` field (see the client configuration examples below).

<Warning>
  Keep your API key private. Do not commit it to public repositories or expose it in plain text in shared client configuration files.
</Warning>

***

## Client Configuration

<Tabs>
  <Tab title="Claude Desktop">
    Add the following to `claude_desktop_config.json`:

    ```json theme={null}
    {
      "mcpServers": {
        "finevoice": {
          "url": "https://mcp.finevoice.ai/v1/mcp",
          "headers": {
            "Authorization": "Bearer YOUR_API_KEY"
          }
        }
      }
    }
    ```
  </Tab>

  <Tab title="Cursor">
    Add the following to `.cursor/mcp.json` in your project root:

    ```json theme={null}
    {
      "mcpServers": {
        "finevoice": {
          "url": "https://mcp.finevoice.ai/v1/mcp",
          "headers": {
            "Authorization": "Bearer YOUR_API_KEY"
          }
        }
      }
    }
    ```
  </Tab>

  <Tab title="VS Code">
    Add the following to `.vscode/mcp.json`:

    ```json theme={null}
    {
      "servers": {
        "finevoice": {
          "type": "http",
          "url": "https://mcp.finevoice.ai/v1/mcp",
          "headers": {
            "Authorization": "Bearer YOUR_API_KEY"
          }
        }
      }
    }
    ```
  </Tab>
</Tabs>

After configuring, restart your client. You can then ask the AI to call FineVoice tools directly in conversation (for example, "Convert this text to speech using the voice james").

***

## Available Tools

Each tool on the MCP Server maps to a FineVoice REST API endpoint, with the same parameters and response structure. Asynchronous tasks are polled with the task status tool, just like the REST API.

### Text to Speech / Podcast

| Tool                      | Endpoint                                     | Description                                                                              |
| ------------------------- | -------------------------------------------- | ---------------------------------------------------------------------------------------- |
| Text to Speech            | `POST /v1/audio/speech-synthesis`            | Convert text to speech — see [TTS](/api-reference/endpoint/tts)                          |
| Podcast Script Generation | `POST /v1/audio/podcastgen/{task_id}/script` | Generate a podcast script — see [Podcast Generation](/api-reference/endpoint/podcastgen) |
| Script to Speech          | `POST /v1/audio/script-tts`                  | Convert a multi-speaker script into speech                                               |

### Speech to Text

| Tool           | Endpoint             | Description                                                                      |
| -------------- | -------------------- | -------------------------------------------------------------------------------- |
| Speech to Text | `POST /v1/audio/stt` | Transcribe audio and generate subtitles — see [STT](/api-reference/endpoint/stt) |

### Sound Effects & Audio Separation

| Tool                    | Endpoint                        | Description                                                                                    |
| ----------------------- | ------------------------------- | ---------------------------------------------------------------------------------------------- |
| Sound Effect Generation | `POST /v1/audio/sfx-generation` | Generate sound effects — see [Sound Effects](/api-reference/endpoint/sfx-generation)           |
| Audio Separation        | `POST /v1/audio/separation`     | Separate vocals and instrumentals — see [Audio Separation](/api-reference/endpoint/separation) |

### Voices / Voice Cloning

| Tool              | Endpoint                          | Description                                                                                           |
| ----------------- | --------------------------------- | ----------------------------------------------------------------------------------------------------- |
| List Voices       | `GET /v1/voice/pagevoices`        | List available voices with pagination — see [Get Voices](/api-reference/endpoint/pagevoices)          |
| Get Voice Detail  | `GET /v1/voice/{voice}`           | Get details for a single voice — see [Get Voice Detail](/api-reference/endpoint/getvoicedetail)       |
| Train Voice Model | `POST /v1/voice/train`            | Train a custom voice clone — see [Train Model](/api-reference/endpoint/trainmodel)                    |
| Design Voice      | `POST /v1/voice/design`           | Design a new voice from a text description — see [Voice Design](/api-reference/endpoint/voice-design) |
| Get Model Status  | `GET /v1/models/{name}`           | Check voice model training status — see [List Models](/api-reference/endpoint/listmodels)             |
| Voice Conversion  | `POST /v1/audio/voice-conversion` | Convert audio to a different voice — see [Voice Conversion](/api-reference/endpoint/voice-conversion) |

### Music Generation

| Tool                                  | Endpoint                                    | Description                                                                                                       |
| ------------------------------------- | ------------------------------------------- | ----------------------------------------------------------------------------------------------------------------- |
| Music Generation                      | `POST /v1/music/musicgen`                   | Generate music — see [Music Generation](/api-reference/endpoint/music-gen)                                        |
| Music Generation by Prompt            | `POST /v1/music/musicgenbyprompt`           | Generate music from a text prompt — see [Music Generation by Prompt](/api-reference/endpoint/music-gen-by-prompt) |
| Music Generation by Structured Prompt | `POST /v1/music/musicgenbystructuredprompt` | Generate music from a structured prompt                                                                           |
| Get Lyrics                            | `GET /v1/music/{music_id}/lyrics`           | Retrieve lyrics for a generated song                                                                              |
| List Artists                          | `GET /v1/music/artists`                     | List available artist styles                                                                                      |
| Music Cover                           | `POST /v1/music/cover`                      | Re-sing a track with a different voice or style — see [Music Cover](/api-reference/endpoint/music-cover)          |

### Audio Enhancement

| Tool                    | Endpoint                                    | Description                                                                                                          |
| ----------------------- | ------------------------------------------- | -------------------------------------------------------------------------------------------------------------------- |
| Speech Enhancement      | `POST /v1/enhancer/speech_enhancement`      | Denoise and enhance speech — see [Speech Enhancement](/api-reference/endpoint/speech-enhancement)                    |
| Speech Super Resolution | `POST /v1/enhancer/speech_super_resolution` | Upsample speech audio quality — see [Speech Super Resolution](/api-reference/endpoint/speech-super-resolution)       |
| Remove Mouth Sounds     | `POST /v1/enhancer/remove_mouth_sounds`     | Remove mouth noises — see [Remove Mouth Sounds](/api-reference/endpoint/remove-mouth-sounds)                         |
| Remove Long Silences    | `POST /v1/enhancer/remove_long_silences`    | Trim long silences — see [Remove Long Silences](/api-reference/endpoint/remove-long-silences)                        |
| Detect Filler Words     | `POST /v1/enhancer/filler_words/detect`     | Detect filler words — see [Filler Words Detect](/api-reference/endpoint/filler-words-detect)                         |
| Remove Filler Words     | `POST /v1/enhancer/filler_words/remove`     | Remove filler words — see [Filler Words Remove](/api-reference/endpoint/filler-words-remove)                         |
| Remove Stuttering       | `POST /v1/enhancer/stuttering/remove`       | Remove stuttered repetitions — see [Stuttering Remove](/api-reference/endpoint/stuttering-remove)                    |
| Audio Normalization     | `POST /v1/enhancer/normalization`           | Normalize audio volume — see [Audio Normalization](/api-reference/endpoint/audio-normalization)                      |
| Enhancer Pipeline       | `POST /v1/enhancer/pipeline`                | Chain multiple enhancement steps — see [Enhancer Pipeline](/api-reference/endpoint/enhancer-pipeline)                |
| Enhancer Separation     | `POST /v1/enhancer/speech_separation`       | Voice separation within the enhancer module — see [Enhancer Separation](/api-reference/endpoint/enhancer-separation) |

### Scoring & Other

| Tool              | Endpoint                                 | Description                                     |
| ----------------- | ---------------------------------------- | ----------------------------------------------- |
| CLAP Score        | `POST /clap-feign-service/v1/clap/score` | Score how well audio matches a text description |
| CLAP Health Check | `GET /clap-feign-service/v1/clap/health` | Health check for the scoring service            |

### Task Polling

| Tool            | Endpoint                 | Description                                                                                      |
| --------------- | ------------------------ | ------------------------------------------------------------------------------------------------ |
| Get Task Status | `GET /v1/task/{task_id}` | Poll the result of an asynchronous task — see [Task Status](/api-reference/endpoint/task-status) |

***

## Notes

* MCP tool calls share the same quota and billing as the REST API — see [API Pricing](/essentials/api-pricing).
* For asynchronous tasks (such as music generation or voice clone training), poll the result with the **Get Task Status** tool after submitting the request via MCP.